AI and Machine Learning Approaches for Price Forecasting of Pulses and Vegetable Commodities

The fluctuation of prices in agri-horti products, such as pulses and major vegetables like onion, potato, and tomato, creates major hurdles for farmers, traders, administrators, and even consumers. In conventional practices, the forecast techniques fail to identify the intricate patterns of the market, which get affected by seasonality, weather patterns, deficiencies in supply-demand, transportation, and market arrivals. The proposed study focuses on the development of reliable AI-ML based predictive systems having the potential for accurate forecast predictions of prices of commodities. Based on historical prices, market arrivals, climate, and trends, machine learning algorithms such as Random Forest, L-STMS, ARIMA, and XGBoost can be used for predicting prices in the short-term and long-term periods.